Dissertations / Theses on the topic 'Imagerie hyperspectrale – Analyse informatique'
Create a spot-on reference in APA, MLA, Chicago, Harvard, and other styles
Consult the top 50 dissertations / theses for your research on the topic 'Imagerie hyperspectrale – Analyse informatique.'
Next to every source in the list of references, there is an 'Add to bibliography' button. Press on it, and we will generate automatically the bibliographic reference to the chosen work in the citation style you need: APA, MLA, Harvard, Chicago, Vancouver, etc.
You can also download the full text of the academic publication as pdf and read online its abstract whenever available in the metadata.
Browse dissertations / theses on a wide variety of disciplines and organise your bibliography correctly.
Pelletier, Bruno. "Traitement neuronal de l'information hyperspectrale." Toulouse, ENSAE, 2002. http://www.theses.fr/2002ESAE0014.
Full textLefévre, Soizic. "Caractérisation de la qualité des raisins par imagerie." Electronic Thesis or Diss., Reims, 2023. http://www.theses.fr/2023REIMS017.
Full textIdentifying the health conditions of the grapes at harvest time is a major issue in order to produce quality wines. To meet this issue, data are acquired by spectrometry, hyperspectral imaging and RGB imaging on grape samples during harvest.Several pre-treatments adapted to each type of data are applied such as normalization, reduction, extraction of characteristic vectors, and segmentation of useful areas. From an imaging point of view, the reconstruction in false colors of hyperspectral images, far from reality, doesn’t allow to label all the intra-class diversity. On the other hand, the visual quality of RGB imaging enables accurate class labelling. From this labelling, classifiers such as support vector machines, random forests, maximum likelihood estimation, spectral mapping, k-means are tested and trained on labelled bases. Depending on the nature of the data, the most effective is applied to whole images of grape clusters or crates of grapes of several grape varieties from different parcels.The quality indices obtained from RGB image processing are very close to the estimates made by experts in the field
Faivre, Adrien. "Analyse d'image hyperspectrale." Thesis, Bourgogne Franche-Comté, 2017. http://www.theses.fr/2017UBFCD075/document.
Full textThis dissertation addresses hyperspectral image analysis, a set of techniques enabling exploitation of micro-spectroscopy images. Images produced by these sensors constitute cubic arrays, meaning that every pixel in the image is actually a spectrum.The size of these images, which is often quite large, calls for an upgrade for classical image analysis algorithms.We start out our investigation with clustering techniques. The main idea is to regroup every spectrum contained in a hyperspectralimage into homogeneous clusters. Spectrums taken across the image can indeed be generated by similar materials, and hence display spectral signatures resembling each other. Clustering is a commonly used method in data analysis. It belongs nonetheless to a class of particularly hard problems to solve, named NP-hard problems. The efficiency of a few heuristics used in practicewere poorly understood until recently. We give theoretical arguments guaranteeing success when the groups studied displaysome statistical property.We then study unmixing techniques. The objective is no longer to decide to which class a pixel belongs, but to understandeach pixel as a mix of basic signatures supposed to arise from pure materials. The mathematical underlying problem is again NP-hard.After studying its complexity, and suggesting two lengthy relaxations, we describe a more practical way to constrain the problemas to obtain regularized solutions.We finally give an overview of other hyperspectral image analysis methods encountered during this thesis, amongst whomare independent component analysis, non-linear dimension reduction, and regression against a spectrum library
Fasquelle, François. "L'identification d'explosifs par imagerie hyperspectrale : spectromètre par transformation de Fourier aéroporté." Thesis, Université Laval, 2006. http://www.theses.ulaval.ca/2006/24097/24097.pdf.
Full textKhoder, Jihan. "Nouvel Algorithme pour la Réduction de la Dimensionnalité en Imagerie Hyperspectrale." Phd thesis, Université de Versailles-Saint Quentin en Yvelines, 2013. http://tel.archives-ouvertes.fr/tel-00939018.
Full textNoyel, Guillaume. "Filtrage, réduction de dimension, classification et segmentation morphologique hyperspectrale." Phd thesis, École Nationale Supérieure des Mines de Paris, 2008. http://pastel.archives-ouvertes.fr/pastel-00004473.
Full textDelcourt, Jonathan. "Un système intégré d'acquisition 3D multispectral : acquisition, codage et compression des données." Phd thesis, Université de Bourgogne, 2010. http://tel.archives-ouvertes.fr/tel-00578448.
Full textVohl, Dany. "Algorithmes de compression d'images hyperspectrales astrophysiques." Thesis, Université Laval, 2013. http://www.theses.ulaval.ca/2013/30110/30110.pdf.
Full textSpIOMM, the Imaging Fourier Transform Spectrometer of the Observatoire du Mont-Mégantic generates huge files of about 700 MB per file on average, and SITELLE, its successor will generate files of a few GB. Since several files can be generated during an observation night and the astronomers are not always on-site, there is an increasing need for both storage and transmission. To minimize storage space, bandwidth use and transmission time, three data compression techniques are presented. The first two techniques are lossless data compression and the third one is lossy. The lossless techniques give better results than generic techniques that are zip and gzip2, with compression ratios varying from 1:19 : 1 to 1:22 : 1. The lossy technique compresses files up to a 64 : 1 ratio. The effect of the lossy process on the photometric measurements and the spectra analysis is also studied.
Lagrange, Adrien. "From representation learning to thematic classification - Application to hierarchical analysis of hyperspectral images." Thesis, Toulouse, INPT, 2019. http://www.theses.fr/2019INPT0095.
Full textNumerous frameworks have been developed in order to analyze the increasing amount of available image data. Among those methods, supervised classification has received considerable attention leading to the development of state-of-the-art classification methods. These methods aim at inferring the class of each observation given a specific class nomenclature by exploiting a set of labeled observations. Thanks to extensive research efforts of the community, classification methods have become very efficient. Nevertheless, the results of a classification remains a highlevel interpretation of the scene since it only gives a single class to summarize all information in a given pixel. Contrary to classification methods, representation learning methods are model-based approaches designed especially to handle high-dimensional data and extract meaningful latent variables. By using physic-based models, these methods allow the user to extract very meaningful variables and get a very detailed interpretation of the considered image. The main objective of this thesis is to develop a unified framework for classification and representation learning. These two methods provide complementary approaches allowing to address the problem using a hierarchical modeling approach. The representation learning approach is used to build a low-level model of the data whereas classification is used to incorporate supervised information and may be seen as a high-level interpretation of the data. Two different paradigms, namely Bayesian models and optimization approaches, are explored to set up this hierarchical model. The proposed models are then tested in the specific context of hyperspectral imaging where the representation learning task is specified as a spectral unmixing problem
Méteau, Jérémy. "Instrumentation optique pour la caractérisation des tissus : analyse de la complémentarité et des limites techniques de fluorescence hyperspectrale et de Tomographie Optique Cohérente en vue de leur intégration multimodale." Thesis, Besançon, 2014. http://www.theses.fr/2014BESA2041/document.
Full textThe aim of this activity is the development of a mono point imaging fiber system which uses hyperspectral multi-excitation auto fluorescence technique for biological tissues and the study of an Optical Coherence Tomography system like another modality. At first, this report presents the optical properties of biological tissues and the relevant fluorophores for cancerous tumors detection. Secondly, the fluorescence imaging system instrumentation and hyperspectral analysis are presented with in vitro results. The third part presents the "scan free" optical coherence tomography system which is able to image without optical displacement. It's characterized and have interesting functionality like depth dependant dispersion compensation. These both techniques are complementary because they get different kind of information. The information of the first one is about biochemical composition of the tissues and the information of the second one is about the stucture
Zullo, Anthony. "Analyse de données fonctionnelles en télédétection hyperspectrale : application à l'étude des paysages agri-forestiers." Thesis, Toulouse 3, 2016. http://www.theses.fr/2016TOU30135/document.
Full textIn hyperspectral imaging, each pixel is associated with a spectrum derived from observed reflectance in d measurement points (i.e., wavelengths). We are often facing a situation where the sample size n is relatively low compared to the number d of variables. This phenomenon called "curse of dimensionality" is well known in multivariate statistics. The mored increases with respect to n, the more standard statistical methodologies performances are degraded. Reflectance spectra incorporate in their spectral dimension a continuum that gives them a functional nature. A hyperspectrum can be modelised by an univariate function of wavelength and his representation produces a curve. The use of functional methods allows to take into account functional aspects such as continuity, spectral bands order, and to overcome strong correlations coming from the discretization grid fineness. The main aim of this thesis is to assess the relevance of the functional approach in the field of hyperspectral remote sensing for statistical analysis. We focused on the nonparametric fonctional regression model, including supervised classification. Firstly, the functional approach has been compared with multivariate methods usually involved in remote sensing. The functional approach outperforms multivariate methods in critical situations where one has a small training sample size combined with relatively homogeneous classes (that is to say, hard to discriminate). Secondly, an alternative to the functional approach to overcome the curse of dimensionality has been proposed using parsimonious models. This latter allows, through the selection of few measurement points, to reduce problem dimensionality while increasing results interpretability. Finally, we were interested in the almost systematic situation where one has contaminated functional data. We proved that for a fixed sample size, the finer the discretization, the better the prediction. In other words, the larger dis compared to n, the more effective the functional statistical methodis
Chavanne, Clarisse. "Les couleurs de la Tapisserie de Bayeux." Electronic Thesis or Diss., Sorbonne université, 2022. http://www.theses.fr/2022SORUS527.
Full textThe Bayeux Tapestry, deposited by the French State in the Bayeux Museum, is a unique 11th century masterpiece listed as a UNESCO World Heritage since 2007. It relates the conquest of England in 1066 by Guillaume the Conqueror, Duke of Normandy, on a canvas almost 70 m long. This work of art, wrongly called tapestry, is actually an embroidery of dyed wool threads on linen. Although it is a subject of study extremely studied by historians and art historians, few researches relating to the scientific analysis of this work, as a textile object, have been carried out. Natural dyes were identified as coming from four plants (madder, weld, woad and an unidentified one producing brown) thanks to invasive analysis in 1982. The objective of this PhD was therefore to obtain more complete information on the dyes used, their degradation but also on the dyeing methods of medieval times, thanks to non-invasive methods. Indeed, no sampling or destructive method is currently possible for works of great importance. Thanks to hyperspectral imaging, we first developed a non-invasive scientific method allowing the identification of the dyes present on the Bayeux Tapestry. For this, we created a database of reflectance spectra of wool dyed then degraded in the laboratory, which we compared to the spectra of the Tapestry. Moreover, by combining the data from the images of the reverse side of the Tapestry with the digital grouping of the dyed threads of the front using the hyperspectral imaging data, it was possible to visually recover the original colours of the Tapestry. The results obtained modify the visual perception and reception of the Bayeux Tapestry by improving the understanding of medieval tastes and practices. These data will allow restaurateurs to apprehend better its restoration in 2026, and will support new researches on the subject
Ghasemzadeh-Barvarz, Massoud. "Quality control of complex polymer materials using hyperspectral imaging associated with multivariate statistical analysis." Doctoral thesis, Université Laval, 2015. http://hdl.handle.net/20.500.11794/26025.
Full textIn the first step of this work, maleic anhydride grafted polyethylene (MAPE)/hemp fiber composites with different filler concentration between 0 to 60% were analyzed by NIR imaging. Three chemometrics methods including Partial Least Squares (PLS), Multivariate Curve Resolution (MCR) and Multivariate Image Analysis/Multivariate Image Regression (MIA/MIR) were implemented and compared for predicting mechanical properties as well as mapping chemical compositions and material physical variations. In the second part of this project, a series of complex multilayer films (industrial samples) were investigated using NIR hyperspectral imaging and Multivariate Image Analysis (MIA). The potential and effectiveness of the proposed method for detecting defects caused by missing layer(s) were studied for unprinted and printed films. In the final part, polypropylene (PP)/glass/flax hybrid composites were studied. The effect of composition (fiber content between 0 and 40 vol%), as well as water, thermal and UV aging on mechanical properties was investigated. Then, Scanning Electron Microscopy (SEM) images of unaged hybrid composites were analyzed using texture analysis and multivariate methods. Finally, NIR imaging was implemented on unaged and aged composites to predict their final properties as well as to capture chemical composition and aging variations. In summary, the results show that combining hyperspectral imaging and multivariate statistical methods is a powerful tool for quality control of complex polymer composites. Since the proposed method is rapid and nondestructive, one can implement it for on-line quality control of finished plastic products. On the other hand, texture analysis and multivariate methods can be applied for quantifying the mixing aspects of composites.
Dary, Christophe. "Analyse géométrique d'image : application à la segmentation multi-échelle des images médicales." Nantes, 1992. http://www.theses.fr/1992NANT07VS.
Full textChen, Jie. "Indentification de système sous la contrainte de non-négativité - Applications dans filtrage adaptatif et analyse dimage hyperspectrale." Phd thesis, Université de Technologie de Troyes, 2013. http://tel.archives-ouvertes.fr/tel-00953563.
Full textFontanilles, Guillaume. "Caractérisation des propriétés optiques infrarouges et de la température par désagrégation de mesures directionnelles et spectrales." Toulouse, ISAE, 2009. http://www.theses.fr/2009ESAE0008.
Full textAmirabadi, Seyedshahabaldin. "Characterization of PLA-Talc films using NIR chemical imaging and multivariate image analysis techniques." Master's thesis, Université Laval, 2018. http://hdl.handle.net/20.500.11794/29853.
Full textFood packaging plays a great role in the food industry to maintain food products quality as long as possible. Polymer films are widely used in food packaging and also attract attention because of their outstanding advantages. Since petroleum-based polymers are known to cause environmental concerns, biodegradable polymers like PLA were studied more intensively in recent years due to their environmentally friendly properties. The application of reinforced films exceeds simple ones in achieving specific functions and enhancing their properties such as barrier properties. Since the films structures are more complex, quality control is more challenging. In industry, off-line methods are vastly used for quality control of the produced films while variability in raw materials and processing conditions substantially change the film specifications. Consequently, real-time inspection and monitoring of polymer films is needed on the production line to achieve a real-time quality control of the films. A fast and non-invasive near-infrared (NIR) imaging system is proposed to characterize biodegradable polylactic acid (PLA) films containing talc, and produced by extrusion film-blowing for packaging applications. The ultimate goal is to use the system for quality control on the extrusion line, and after a post-processing via thermal treatment (annealing). A set of PLA-talc films with varying talc contents were produced and submitted to annealing under different conditions. NIR images of the films were collected after which the samples were characterized for their physical, mechanical, and gas barrier properties. Multivariate imaging techniques were then applied to the hyperspectral images. It is shown that various talc loadings can be distinguished, and the information contained in the NIR spectra allows predicting the film properties. In all cases, the proposed approach was able to track the variation in film properties with good accuracy
Boiret, Mathieu. "Towards chemometric methodologies on hyperspectral imaging for low dose compound detection : application on Raman microscopy." Thesis, Montpellier, 2015. http://www.theses.fr/2015MONTS291.
Full textHyperspectral imaging is now considered as a powerful analytical tool in the pharmaceutical environment, both during development to ensure the drug product quality and to solve production issues on commercialized products.In this thesis, Raman microscopy is used to study the distribution of actives and excipients in a pharmaceutical drug product, by especially focusing on the identification of a low dose compound. This latter product is defined as a compound which has low spatial and spectra contributions, meaning that it is scattered in a few pixels of the image and that its spectral response is mixed with the other compounds of the formulation. While most chemometric tools are based on the decomposition of statistical moments (requiring sufficient variations between samples or image pixels), some limitations have been rapidly reached. The first part of this thesis highlights the difficulty to detect a low dose compound in a product by using independent component analysis or multivariate curve resolution. Different methodologies are proposed to circumvent these limitations. For both techniques, reduction of dimensions and filtering steps appears as critical parameters of the method. The second part of the thesis focusses on the signal space to determine absence/presence compound maps or to detect the compounds in an unknown formulation. The proposed methods are only based on the spectral space of each formulation compound. There are perfectly suitable to a low dose compound and should be well-adapted to other analytical techniques or to other environments
Guignard, Léo. "Analyse quantitative de la morphogenèse animale : de l'imagerie laser haut-débit à l'embryon virtuel chez les ascidies." Thesis, Montpellier, 2015. http://www.theses.fr/2015MONTS048/document.
Full textAscidian embryos develop with stereotyped and evolutionarily conserved invariant cell lineages to produce in a few hours or days tadpole larvae with a small number of cells. They thus provide an attractive framework to describe with cellular resolution the developmental program of a whole organism. During my PhD, I developed a quantitative approach to describe the evolution of embryonic morphologies during the development of the ascidian Phallusia mammillata. I then used this approach to systematically characterize in detail the logic of cell fate induction events. To quantitatively characterize cell behaviors during embryogenesis, we used multi-angle light-sheet microscopy to image with high spatio-temporal resolution entire live embryos with fluorescently labeled plasma membranes. To extract biological information from this imaging dataset, I then developed a conceptually novel automated method for 4D cell segmentation, ASTEC. Applied to a Phallusia mammillata embryo imaged for 6 hours between the 64-cell and the initial tailbud stages, this method allows the accurate tracking and shape analysis of 1030 cells across 640 cell divisions. The resulting 4D digital embryo can be formalized as a dynamic graph, in which cells are represented by nodes, linked within a time point by edges that represent their spatial neighborhood, and between time points by temporal edges describing cell lineages.Based on this quantitative digital representation, we systematically identified cell fate specification events up to the late gastrula stage. Computational simulations revealed that remarkably simple rules integrating measured cell-cell contact areas with boolean spatio-temporal expression data for extracellular signalling molecules are sufficient to explain most early cell inductions. This work suggests that in embryos establishing precise stereotyped contacts between neighboring cells, the genomic constraints for precise gene expression levels are relaxed, thereby allowing rapid genome evolution
Rolland, Yan. "Elements de perception , de modelisation et d'analyse de texture (doctorat : biologie et sciences de la sante)." Rennes 1, 1997. http://www.theses.fr/1997REN1B022.
Full textJuliachs, Manuel. "Rendu volumique parallèle hybride de maillages non structurés." Versailles-St Quentin en Yvelines, 2008. http://www.theses.fr/2008VERS0016.
Full textIn recent years, performance and functionalities of graphics processing units (or GPUs) have tremendously increased, allowing the development of novel algorithms in the field of unstructured meshes volume rendering for scientific visualization. In this thesis report, we present our contributions to this particular domain. First, after reviewing parallel graphics rendering as well as the real-time graphics pipeline, we conduct a bibliographical survey of the previous work regarding unstructured meshes volume rendering on the GPU. Following this, we present a GPU point-sprite based tetrahedral meshes volume rendering algorithm. It minimizes the amount of data transmitted to the GPU for each tetrahedron rendered, resulting in a lower bandwidth and a higher performance. Afterwards, we present a hybrid parallel tetrahedral meshes volume rendering architecture. It partitions its work between the CPU and the GPU, according to the space (object or image) into which each work subset is done, allowing to arbitrarily scale the number of nodes used. Finally, we discuss how our two main contributions, namely point-sprite based and hybrid parallel volume rendering, might be adapted to future hardware architectures in order to improve both their performance and rendering quality
Foucher, Christophe. "Analyse et amélioration d'algorithmes neuronaux et non neuronaux de quantification vectorielle pour la compression d'images." Rennes 1, 2002. http://www.theses.fr/2002REN10120.
Full textRousseau, Batiste. "Étude de la composition et des propriétés physiques de surface de la comète 67P/Churyumov-Gerasimenko : interprétation des données VIRTIS/Rosetta etmesure en réflectance d’analogues cométaires." Thesis, Paris Sciences et Lettres (ComUE), 2017. http://www.theses.fr/2017PSLEO018/document.
Full textDuring the Solar System formation, 4.6 billion years ago, comets accreted materials which have been transformed according to the physical and dynamical conditions of the accretion disk but also a part of components coming from the interstellar medium. By preserving a primordial composition, the study of comets allows us to better understand the conditions of the proto-planetary disk surrounding the young Sun of an epoch which is now inaccessible. Moreover, it consists also to understand the various comets populations, their formation process, dynamical and activity evolution as they inward and outward the Sun or their structure.The ESA/Rosetta mission followed the comet 67P/Churyumov-Gerasimenko during two years. A ten of instruments has been dedicated to the study of the evolution of its activity, gas release, surface morphology, dust and other objectives. VIRTIS is a visible/infrared spectrometer instrument. It is composed of VIRTIS-M, an imaging spectrometer which gives access to spatial information with moderate spectral resolution and VIRTIS-H, a point spectrometer with a higher spectral resolution. This study is based on the data analysis of VIRTIS instruments and is divided into two parts focused on the study of the nucleus surface.The first part is an analysis of the spectral and photometric parameters: albedo, spectral slope, the main direction of the light diffusion by particles, macroscopic roughness. In a global study, I highlighted the spatial variations of albedo and spectral slope; compared results derived from different models as well as from both instruments. Then, I determined these parameters locally, revealing differences between two types of terrains. This approach allows to better understand the mechanisms linked to the activity (dust drop-off/uprising, space weathering, ice content variation) and also to the surface properties (composition, texture).The second goal of the thesis is to reproduce in the laboratory the observations realized by VIRTIS to give constraints on the composition and texture of the surface. In collaboration with IPAG (Grenoble, France) I led experiments consisting of the production of very fine powders made of materials which look like those we suspect to be present on the nucleus of 67P: organic matter (mimicked by a coal), silicates (olivine) and iron sulfides (pyrite and pyrrhotite) are all observed on comets or their analogues. I ground them to micrometric to nanometric scales and I realized reflectance measurements in the same spectral range than VIRTIS. Then, I have been able to observe effects caused by the variations of the grain size, composition or texture of the mixture and to highlight combinations reproducing the mean comet VIRTIS spectrum. Finally, this work enables us understanding the influence of material poorly studied such as iron sulfides as well as the spectral behaviour of powders composed of grain sizes reaching an order of magnitude close to the wavelengths, which is essential in the study of cometary surfaces
Abou, Fadel Maya. "Apports de la chimiométrie à la spectroscopie de Résonance Paramagnétique Electronique : nouvelles perspectives de traitement de données spectrales à dimensions spatiales (imagerie) et/ou temporelles." Thesis, Lille 1, 2015. http://www.theses.fr/2015LIL10130/document.
Full textElectron Paramagnetic Resonance (EPR) Spectroscopy has undoubtedly become the first-choice technique for the characterization of complex materials containing some unpaired electrons (transition metal ions, rare earth ions, defects, organic radicals ...). Similarly to nuclear magnetic resonance spectroscopy, EPR generates multidimensional (2D, 3D…) spectral and recently also spatial (imaging) data as well as spectral/spatial ones. It is thus, surprising that despite the large amount of spectral data to be explored and complexity of the EPR signals, there hardly exist at the international level of exploitation the multivariate data processing methods that are widely available in chemometrics. The objective of this thesis is thus, to develop new tools for the treatment of these EPR spectral data, to establish new analytical methodologies and to evaluate their performance. The two main areas that will be studied are spectroscopic imaging and time-resolved spectroscopy. In this work, we will show that the implementation of the methods known as "multivariate curve resolutions" can extract, simultaneously, and without a priori all chemical maps and their corresponding spectra of pure compounds present in the studied sample. This methodology will also be exploited to extract the EPR spectra of intermediate species during a kinetic monitoring
Laguitton, Soizic. "Suivi spatio-temporel des artères coronaires en imagerie scanner multidétecteur." Rennes 1, 2008. http://www.theses.fr/2008REN1S003.
Full textThe application we are dealing with concerns the image guided endovascular surgery. Our aim is to assess the optimal catheter trajectory, by analysing the surroundings from the inside of the deformable structures in motion. The strategy is to track the coronary in 3D+T MSCT sequences. It researches the displacement of a point located on the vessel during a cardiac cycle. A coarse research is first performed and then refined. The point selection criterion uses Euclidean distance between local descriptors of the vessel. The accuracy and robustness of the developed methods are measured on sequences containing simulated linear and non linear movements. We applied the methods to extract trajectories of coronary segments during one cardiac cycle
Jocelyne, Fayn. "L'Analyse séquentielle des électrocardiogrammes." Lyon, INSA, 1990. http://www.theses.fr/1990ISAL0007.
Full textMangin, Franck. "Amélioration de la détection de contours en imagerie artificielle par un modèle coopératif multi-résolution." Nice, 1994. http://www.theses.fr/1994NICE4715.
Full textDebon, Renaud. "Analyse d'images échographiques de loesophage, reconstruction 3D et interprétation." Rennes 1, 2005. http://www.theses.fr/2005REN1S196.
Full textThis work concerns the approach evaluations, which are oriented « knowledge based system » for medical images interpretation applied to esophagus echoendoscopy and the development of aid system for tumor staging. It's shown how the intelligent approaches (expert system and information fusion) can allow rationalizing the using of a priori knowledge. The pertinent extraction of anatomic structures, in our case, esophagus structure, becomes a natural application in the knowledge engineering. This extraction is based on the image segmentation. The required robustness for these algorithms impose the advanced architectures development allowing the compensation of low numerical content of these images. Three concrete examples are detailed : 2D extraction of the esophagus' interface, 3D extraction of the esophagus' interfaces and spatial following with 3D reconstruction of the aorta. Knowledge is represented by static or dynamical model (fuzzy, geometric or evidential models). The approach using the complementarities of probabilities and fuzzy logic to obtain presentation « exact » of knowledge a priori. Fuzzy models and statistic reality are synchronized by a knowledge base. It's shown that all components can be integrated in a coherent architecture hierarchically organized
Samir, Chafik. "Analyse des déformations des visages 3D utilisant les chemins géodésiques dans l'espace des surfaces faciales." Evry, Institut national des télécommunications, 2006. http://www.theses.fr/2007TELE0016.
Full text[non communiqué]
Parra-Denis, Estelle. "Analyse morphologique 3D de particules de formes complexes : application aux intermétalliques dans les alliages d'aluminium." Saint-Etienne, 2007. http://www.theses.fr/2007STET4001.
Full textIn this PhD-thesis, the studied material is an aluminium alloys 5182 containing 2 particles types : Alx(Fe, Mn) and Mg2Si. During industrial rolling, particles are broken up, and redistributed in the volume of the sheet. As particles controls surface aspect and formability of the final sheet, the work presented is based on 2 issues : to highlight volume distribution of intermetallic particles, and to characterize their three-dimensional shapes during the rolling process. Thus, samples extracted to different steps of the rolling process are been scanned by X ray microtomography (performed at the ESRF). The obtained three dimensional images are segmented by a multi class thresholding method. Extraction of each particle type is allowed. A data basis containing each individualized particles is created. It ensures to reconstruct the materail binary image and also to work particle by particle. The particles dispersion is highlight by the graph of distances and the covariance study. The analysis of linear erosion and covariance allows to model the material by a probabilistic Boolean model with spherical primary grains. A set of morphological parameters characterizes the intermetallic particles three dimensional complex shapes. A principal components analysis is performed to sum up the information. In the obtained subspace, particles are classified into 5 families of shapes. Their evolution is followed during the rolling process
Boulais, Axel. "Méthodes de séparation aveugle de sources et application à l'imagerie hyperspectrale en astrophysique." Thesis, Toulouse 3, 2017. http://www.theses.fr/2017TOU30318/document.
Full textThis thesis deals with the development of new blind separation methods for linear instantaneous mixtures applicable to astrophysical hyperspectral data sets. We propose three approaches to perform data separation. A first contribution is based on hybridization of two existing blind source separation (BSS) methods: the SpaceCORR method, requiring a sparsity assumption, and a non-negative matrix factorization (NMF) method. We show that using SpaceCORR results to initialize the NMF improves the performance of the methods used alone. We then proposed a first original method to relax the sparsity constraint of SpaceCORR. The method called MASS (Maximum Angle Source Separation) is a geometric method based on the extraction of single-source pixels to achieve the separation of data. We also studied the hybridization of MASS with the NMF. Finally, we proposed an approach to relax the sparsity constraint of SpaceCORR. The original method called SIBIS (Subspace-Intersection Blind Identification and Separation) is a geometric method based on the identification of intersections of subspaces generated by regions of the hyperspectral image. Under a sparsity assumption, these intersections allow one to achieve the separation of the data. The approaches proposed in this manuscript have been validated by experimentations on simulated data and then applied to real data. The results obtained on our data are very encouraging and are compared with those obtained by methods from the literature
Boukarri, Bachir. "Reconstruction 3D récursive de scènes structurées au moyen d'une caméra mobile : application à la robotique." Paris 11, 1989. http://www.theses.fr/1989PA112290.
Full textHatt, Mathieu. "Analyse et traitement d'images multi modales en oncologie." Habilitation à diriger des recherches, Université de Bretagne occidentale - Brest, 2012. http://tel.archives-ouvertes.fr/tel-00721743.
Full textHanafi, Maher. "Analyse du mouvement humain par vison artificielle pour consoles de jeux vidéos." Thesis, Université Laval, 2012. http://www.theses.ulaval.ca/2012/28483/28483.pdf.
Full textThis report is interested in the markerless motion tracking and the three-dimensional reconstruction of articulated human movements from video sequences acquired with a 3D camera (stereo or infrared). Nowadays, this research field is very active and contains a wide field of applications which deals with areas such as motion capture for animations and virtual reality without using any kind of markers, human-machine interaction (HMI), remote monitoring and of course video games. In this manuscript, we propose a novel method to estimate the 3D human pose. This markerless technique is based on an alignment of a skeleton and a 3D human model over the silhouette seen by the camera by leaning on a progressive adjustment, starting from the head and continuing to the trunk and the various members. The technique considers the articulated aspect of the human body and allows, in particular, solving some problems of occlusions and overlapping. Besides, the complexity of the human body structure, of its physical constraints as well as the big variability in the images’ observations, makes that the solution determination for this problem is difficult. The objective of this memory thus is to develop a strong and robust method capable of facing these various difficulties imposed by the technology choice and the general context of use for home video games consoles. To approach this study, we propose a 3D human model which takes into account physical and kinematic constraints and which allows a coherent integration of various visual information such as face detection, edges and silhouettes. The combined system allows 3D human motion tracking using only one 3D camera.
Baguet, Jean-Philippe. "Analyse structurale et fonctionnelle de la paroi carotidienne par échographie assistée par ordinateur : reconstruction 3D de l'artère carotide." Université Joseph Fourier (Grenoble), 2000. http://www.theses.fr/2000GRE19010.
Full textBanga, Mbom Calvin David. "L'approche bootstrap en analyse des images : application à la restitution de la cinétique de la fuite dans la choriorétinopathie séreuse centrale." Rennes 1, 1995. http://www.theses.fr/1995REN10016.
Full textThis work concerns the medical image analysis using statistical pattern recognition methods. These methods usually lead to classification and multivariage segmentation approaches that are more and more used in medical imaging for qualitative or quantitative analysis of tissue. However, statistical pattern reconigtion methods are well known to need high computing time. This assumption is not suitable for image sequences analysis. To get rid of this important problem, we suggest a bootstrap approach for two-dimensional image analysis. This model is based on the random sampling of blocks of observations in the original image. The correlation relationships are maintained in a given block, while the selected blocks are each other independent. Given an original image, we randomly select a small representative set of observations for image statistical parameters estimation. In this way, the bootstrap model improves the estimation and reduces the computing time due to the complexity of estimation algorithms used in statistical pattern recognition. The problem of the quantification of the serum leak kinetic in the Central Serous Choroiditis pathology is and important application in wich the bootstrap model we propose is fro a great need. The serum leak surface is mesures from the eye's fundus images sequence after an unsupervised segmentation sept using the bootstrap random sampling model wich we have proposed. Both a high-quality segmentation and a great reduction of etimation time are required for this application. A graph showing the serum leak surface versus the Fluorecein inection time is then plotted for helping the decision in ophtalmology. The bootstrap model wich we propose for image analysis shows the way to use the bootstrap approach for statistical parameters estimation in pattern recognition, notably in image analysis by invariance methods, in texture analysis and multivariate segmentation
Pottier, Fabien. "Etude des matières picturales du Codex Borbonicus – Apport des spectroscopies non-invasives à la codicologie." Thesis, Cergy-Pontoise, 2017. http://www.theses.fr/2017CERG0908/document.
Full textThe Codex Borbonicus is a great source of knowledge regarding different aspects of the culture of the basin of Mexico-Tenochtitlan at the time of the Spanish conquest. For some historians, the manuscript is definitely Precolumbian while for others, parts of its graphical contents reveal a European cultural influence. In order to investigate the technological knowledge that was involved for its creation, and to bring fresh data to the debate, the manuscript material constituents are studied with transportable, non-invasive analytical techniques (X-Ray fluorescence, reflexion, emission and Raman spectroscopies). A first interpretation of the analytical data recorded on the document is based on the historical records and the corpus of Mesoamerican manuscripts that have already been studied. A finer analysis of the data is done through the calculation of spectral combinations as well as the experimental studies of certain coloring materials, that allow a better understanding of the paint preparation techniques. In order to generalize the conclusion based on localized analyses, the spatial distribution of the constituent on the whole document is also explored. Hyperspectral imaging, with the aid of statistical tools and the mapping of specific spectral features, brings new insights to the first results. The exclusive use of organic colorants extracted from animal (Dactylopius coccus) or vegetable sources (Indigofera suffruticosa, Comellina coelestis, Justicia spicigera) in the Codex Borbonicus, alone or in mixtures, fits what is known of Precolumbian traditions. Therefore, the hypothesis of a European influence can’t be supported by the nature of the manuscript constituents. Moreover, these new data enrich the current knowledge on Mesoamerican manuscript production techniques
Riyanto. "Simulation, optimisation, et analyse de performances pour les systèmes industriels d'acquisition d'images." Toulouse, INPT, 1997. http://www.theses.fr/1997INPT107H.
Full textLandry, David. "Data-driven covariance estimation for the iterative closest point algorithm." Master's thesis, Université Laval, 2019. http://hdl.handle.net/20.500.11794/34734.
Full textThree-dimensional point clouds are an ubiquitous data format in robotics. They are produced by specialized sensors such as lidars or depth cameras. The point clouds generated by those sensors are used for state estimation tasks like mapping and localization. Point cloud registration algorithms, such as Iterative Closest Point (ICP), allow us to make ego-motion measurements necessary to those tasks. The fusion of ICP registrations in existing state estimation frameworks relies on an accurate estimation of their uncertainty. Unfortunately, existing covariance estimation methods often scale poorly to the 3D case. This thesis aims to estimate the uncertainty of ICP registrations for 3D point clouds. First, it poses theoretical foundations from which we can articulate a covariance estimation method. It reviews the ICP algorithm, with a special focus on the parts of it that are pertinent to covariance estimation. Then, an inserted article introduces CELLO-3D, our data-driven covariance estimation method for ICP. The article contains a thorough experimental validation of the new algorithm. The latter is shown to perform better than existing covariance estimation techniques in a wide variety of environments. Finally, this thesis comprises supplementary experiments, which complement the article.
Reis, Philippe. "Vision monoculaire pour la navigation d'un robot mobile dans un univers partiellement modelisé." Clermont-Ferrand 2, 1991. http://www.theses.fr/1991CLF21297.
Full textFerté, Morgan. "Étude et analyse de couches minces par techniques multi-spectroscopiques pour une application sur une ligne de galvanisation." Thesis, Université de Lorraine, 2014. http://www.theses.fr/2014LORR0206/document.
Full textWith the emergence of new steel loaded in reactive elements, the characterization of the physico-chemical properties of the surface states has taken a new dimension. Indeed, the thin films of oxide formed are damaging for the surface quality of the steel and may adversely affect the application of varied coatings. Also, to ensure a good quality on finished products, the need to characterize, online, the steel surface over its entire width is increasingly necessary. The hyperspectral imaging is an emerging technology that allows both spatial and spectral characterization of a surface. It had never been used to characterize thin layers in the steel industry. During my PhD, both methodologies have been developed to meet this need: one via a measurement in specular reflection and the other via a measurement in emission. In addition to a synthesis of reference samples, a full experimental development ranging from a simulation bench to the data processing was performed. This work has demonstrated the capability of the hyperspectral imaging to detect thin surface oxide layers on a steel surface, sometimes at several hundred degrees Celsius. These encouraging results led to the first industrial implementation of this technology. This thesis made it possible to study theoretically and experimentally the phenomena involved and to move from the concept that meets an industrial need to the implementation of an online sensor dedicated to the characterization of a thin oxide layer on a moving surface in industrial conditions
Triquet, Frédéric Chaillou Christophe Meseure Philippe. "Habillage de modèles mécaniques facettisation temps réel de surfaces implicites /." [S.l.] : [s.n.], 2001. http://www.univ-lille1.fr/bustl-grisemine/pdf/extheses/50376-2001-255-256.pdf.
Full textGevaux, Lou. "3D-hyperspectral imaging and optical analysis of skin for the human face." Thesis, Lyon, 2019. http://www.theses.fr/2019LYSES035.
Full textHyperspectral imaging (HSI), a non-invasive, in vivo imaging method that can be applied to measure skin spectral reflectance, has shown great potential for the analysis of skin optical properties on small, flat areas: by combining a skin model, a model of light-skin interaction and an optimization algorithm, an estimation of skin chromophore concentration in each pixel of the image can be obtained, corresponding to quantities such as melanin and blood. The purpose of this work is to extend this method to large, non-flat areas, in particular the human face. The accurate measurement of complex objects such as the face must account for variances of illumination that result from the 3D geometry of an object, which we call irradiance drifts. Unless they are accounted for, irradiance drifts will lead to errors in the hyperspectral image analysis.In the first part of the work, we propose a measurement setup comprising a wide field HSI camera (with an acquisition range of 400 - 700 nm, in 10 nm width wavebands) and a 3D measurement system using fringe projection. As short acquisition time is crucial for in vivo measurement, a trade-off between resolution and speed has been made so that the acquisition time remains under 5 seconds.To account for irradiance drifts, a correction method using the surface 3D geometry and radiometry principles is proposed. The irradiance received on the face is computed for each pixel of the image, and the resulting data used to suppress the irradiance drifts in the measured hyperspectral image. This acts as a pre-processing step to be applied before image analysis. This method, however, failed to yield satisfactory results on those parts of the face almost perpendicular to the optical axis of the camera, such as the sides of the nose, and was therefore discarded in favor of using an optimization algorithm robust to irradiance drifts in the analysis method.Skin analysis from the measured hyperspectral image is performed using optical models and an optimization method. Skin is modeled as a two-layer translucent material whose absorption and scattering properties are determined by its composition in chromophores. Light-skin interactions are modeled using a two-flux method. An inverse problem is solved by optimization to retrieve information about skin composition from the measured reflectance. The chosen optical models represent a trade-off between accuracy and acceptable computation time, which increases exponentially with the number of parameters in the model. The resulting chromophore maps can be added to the 3D mesh measured using the 3D-HSI camera for display purposes.In the spectral reflectance analysis method, skin scattering properties are assumed to be the same for everyone and on every part of the body, which represents a shortcoming. In the second part of this work, the fringe projector originally intended for measuring 3D geometry is used to acquire skin modulation transfer function (MTF), a quantity that yields information about both skin absorption and scattering coefficients. The MTF is measured using spatial frequency domain imaging (SFDI) and analyzed by an optical model relying on the diffusion equation to estimate skin scattering coefficients. On non-flat objects, retrieving such information independently from irradiance drifts is a significant challenge. The novelty of the proposed method is that it combines HSI and SFDI to obtain skin scattering coefficient maps of the face independently from its shape.We emphasize throughout this dissertation the importance of short acquisition time for in vivo measurement. The HSI analysis method, however, is extremely time-consuming, preventing real time image analysis. A preliminary attempt to address this shortcoming is presented, using neural networks to replace optimization-based analysis. Initial results of the method have been promising, and could drastically reduce calculation time from around an hour to a second
Al, Hayek Marianne. "Modélisation optique de signatures spectrales et polarimétriques d'objets pour augmenter les performances d'un système de reconnaissance." Electronic Thesis or Diss., Brest, 2023. http://www.theses.fr/2023BRES0101.
Full textConventional imaging, limited to object shapes and colors, faces limitations in object recognition. To enhance imaging system performance, hyperspectral and polarimetric imaging provides a wealth of information, includingchallenging-to-obtain physical parameters. This facilitates improved object detection, quantitative characterization, and classification. However, the processing of complex data from these modalities remains a challenge. The aim of this work is to propose a generic methodology for the analysis of optical signals, with a primary focus on hyperspectral imaging (HSI). An original classification of invertible physics-based hyperspectral models is presented, along with descriptions of recent diverse models for various applications: MPBOM for algae and bacteria biofilm, MARMIT for soil, PROSPECT for plant leaves, Farrell for turbid biological tissues, Schmitt for human skin, and Hapke for objects in the solar system. A convergence between the PROSPECT and Farrell models for intermediate objects (green apple and leek) paves the way for the development of a new generic and comprehensive modeling approach.Particularly in the field of biology, in collaboration with the ANSES laboratory, we conducted early detection ollowed by quantification of biofilms forming in fish farming basins using hyperspectral and polarimetric imaging. This is crucial as the current visual detection method is not efficient in preventing biofilm accumulation and implementingcleaning and disinfection procedures. Hence, an initial version of a dedicated physical modeling approach called "DNA-HSI" has been established
Feydy, Jean. "Analyse de données géométriques, au delà des convolutions." Thesis, université Paris-Saclay, 2020. http://www.theses.fr/2020UPASN017.
Full textGeometric data analysis, beyond convolutionsTo model interactions between points, a simple option is to rely on weighted sums known as convolutions. Over the last decade, this operation has become a building block for deep learning architectures with an impact on many applied fields. We should not forget, however, that the convolution product is far from being the be-all and end-all of computational mathematics.To let researchers explore new directions, we present robust, efficient and principled implementations of three underrated operations: 1. Generic manipulations of distance-like matrices, including kernel matrix-vector products and nearest-neighbor searches.2. Optimal transport, which generalizes sorting to spaces of dimension D > 1.3. Hamiltonian geodesic shooting, which replaces linear interpolation when no relevant algebraic structure can be defined on a metric space of features.Our PyTorch/NumPy routines fully support automatic differentiation and scale up to millions of samples in seconds. They generally outperform baseline GPU implementations with x10 to x1,000 speed-ups and keep linear instead of quadratic memory footprints. These new tools are packaged in the KeOps (kernel methods) and GeomLoss (optimal transport) libraries, with applications that range from machine learning to medical imaging. Documentation is available at: www.kernel-operations.io/keops and /geomloss
Dubé, Julie. "Analyse de la reconstruction 3D par stéréo multivue dans l'optique des défis de l'appariement." Thesis, Université Laval, 2009. http://www.theses.ulaval.ca/2009/26248/26248.pdf.
Full textEskandar, Fadi. "Informatique et Imagerie Médicale. Rôle de l’INRIA des années 1970 à nos jours. Exemple de l’équipe-projet EPIDAURE/ASCLEPIOS." Thesis, Sorbonne université, 2019. http://www.theses.fr/2019SORUL143.
Full textToday, medical imaging is a major component of the medicine development: image analysis and its deployment are a central research area at INRIA (French Institute for Research in Computer Science and Automation), pioneer in this field, in France and abroad. However, it is difficult to evaluate this kind of fundamental research because of its long-term realization. Moreover, given that the research at INRIA is based on cooperation, its role may be underestimated for the benefit of its partners. The historical objective of this thesis is to show the importance of INRIA’s role in medical imaging analysis through the example of EPIDAURE / ASCLEPIOS, leading team in this field. This investigation is based on three methodological paradigms: a bibliographic search, a synthesis from the archives of INRIA-Rocquencourt (Ile-de-France) and ten semi-directed interviews, conducted by us, with INRIA researchers directly involved in Medical imaging the analysis. The results show that since the 90s of the 20th century, the EPIDAURE / ASCLEPIOS project-team has been the founder of this field in the world. Moreover, this team is the origin of a new digital revolution, through the design of the "Personalized Digital Patient", proposed by its director Nicholas AYACHE, whose central role in this line of research is reflected in the excellence of published results, the wealth and dynamism of established collaborations in the academic, medical and industrial fields. The support of INRIA’s management to the EPIDAURE / ASCLEPIOS project-team is an asset of its success on the international scene
Lescoat, Thibault. "Geometric operators for 3D modeling using dictionary-based shape representations." Electronic Thesis or Diss., Institut polytechnique de Paris, 2020. http://www.theses.fr/2020IPPAT005.
Full textIn this thesis, we study high-level 3D shape representations and developed the algorithm primitives necessary to manipulate shapes represented as a composition of several parts. We first review existing representations, starting with the usual low-level ones and then expanding on a high-level family of shape representations, based on dictionaries. Notably, we focus on representing shapes via a discrete composition of atoms from a dictionary of parts.We observe that there was no method to smoothly blend non-overlapping atoms while still looking plausible. Indeed, most methods either required overlapping parts or do not preserve large-scale details. Moreover, very few methods guaranteed the exact preservation of the input, which is very important when dealing with artist-authored meshes to avoid destroying the artist's work. We address this challenge by proposing a composition operator that is guaranteed to exactly keep the input while also propagating large-scale details.To improve the speed of our composition operator and allow interactive edition, we propose to simplify the input parts prior to completing them. This allow us to interactively previsualize the composition of large meshes. For this, we introduce a method to simplify a detailed mesh to a coarse one by preserving the large details. While more constrained than related approaches that do not produce a mesh, our method still yields faithful outputs
Petit, Tristan. "Caractérisation des fonds marins et estimation bathymétrique par inversion de modèle de transfert radiatif : application à l'imagerie hyperspectrale en milieu coralien." Thesis, Brest, 2017. http://www.theses.fr/2017BRES0023/document.
Full textAirborne hyperspectral imaging is a potential candidate for mapping and monitoring coral reefs at large scale and with high spatial resolution. In this thesis, we first present the processing steps to be applied to hyperspectral signals for extracting information about seabed types, bathymetry and water optical properties, and we discuss their efficiency with respect to two main confounding factors: (i) low signal to noise ratio of the measured signals, and (ii) large number and variability of physical interactions occurring between the entrance of sunlight into the atmosphere and its measurement by the hyperspectral sensor. Considering these limitations, we examine the performance of an already existing water column processing method: semi-analytical model inversion by optimization. We first evaluate the robustness of seabed type and bathymetry estimation for six different inversion setups. The results on hyperspectral images acquired over Réunion Island reefs in 2009 show that the choice of the inversion setup plays an important role on the quality of the estimations and that the most widely used inversion setup does not always produce the best results. We then evaluate the importance of the accuracy of the parameterization of the direct semi-analytical model. This is done through a sensitivity analysis performed on both simulated and real hyperspectral data acquired in Réunion Island in 2015. The analysis is performed for each inversion setup previously studied. This study shows that in coral reef context the accuracy of the parameterization of the direct model is less important than the choice of the inversion setup. We also demonstrate that it is not possible to identify the most influent parameters of the direct model because it depends on the relative concentration of each optically active constituent
Triquet, Frédéric. "Habillage de modèles mécaniques : facettisation temps réel de surfaces implicites." Lille 1, 2001. https://pepite-depot.univ-lille.fr/LIBRE/Th_Num/2001/50376-2001-255-256.pdf.
Full text